Hi, On Sun, Apr 1, 2012 at 3:53 AM, Angadh Nanjangud <angad...@gmail.com> wrote: > > Gilbert's latest code does use scipy but our thoughts are to eliminate > external dependencies like that and just have the solver in sympy. > I do intend to look into getting code output for different platforms. > But I was also thinking about maybe writing a Runge-Kutta (or other > commonly used numerical method) solver in SymPy. Any thoughts on that?
The problem I see is that any numerical method written in SymPy for solving ODEs will be much slower than using SciPy. I don't see the point in duplicating functionality that's readily available (and is better) elsewhere. The advantages I see in a SymPy numerical ODE method are a) that it can use arbitrary precision and could support more precise tolerances and b) that it could be used in a pure Python environment. I don't see that these are particularly important for numerical ODE solving, but others may differ. I'd rather see effort put into improving the interaction with external ODE solvers (however that may be done). Part of the problem is that writing numerical ODE solvers can be tricky to do well (but it's not hard to do poorly). There's a reason people are still using old Fortran solvers rather than rewriting them. Cheers, Tim. -- Tim Lahey PhD Candidate, Systems Design Engineering University of Waterloo http://about.me/tjlahey -- You received this message because you are subscribed to the Google Groups "sympy" group. To post to this group, send email to sympy@googlegroups.com. To unsubscribe from this group, send email to sympy+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/sympy?hl=en.
